Wildfire, Historical, Baseline, US and Canada (v3.0)

Technical Description

The Wildfire Historical Risk product captures the recent near-historical risk of wildfires across the US and Canada using well-established fire risk indices, as well as a custom vegetative fuel load dataset.

This layer displays the historical fire risk or the likelihood of fire occurrence in the past decade. It does not predict wildfires, but identifies regions which, due to their localized climate and vegetation characteristics, are more likely to experience wildfires based on a significant period of near-historical data. It consists of risk scores calculated using a combination of historical Fire Weather Index (FWI) values between the years 2010-2021 and a static vegetative fuel layer. Historical wildfire events are not used in the creation of the model, but they were used in the course of validation to ensure areas of observed fire corresponded to elevated risk values.

The FWI is an index which indicates fire danger due to climate factors such as temperature, wind and precipitation. An extensive record of historical fire weather information is used to identify anomalously high fire weather conditions which are scaled, corrected and combined with our proprietary fuel layer. The values range from 0 - 100 representing low to extreme fire risk. All croplands/agricultural areas in the US and Canada have been masked in this layer. Additionally, urban cores have also been masked, but fire risk is still calculated at the urban-wildland interface.

This dataset contains the following fields:

• risk: The wildfire risk score at that location. This is a value in the range 0-100.
• fire_risk_class: The fire risk classification for each value. Values are categorized into a 1-5 risk classification system, ranging from Low to Extreme, based on an analysis of historically observed fires in relation to the risk field. The risk classifications are defined as follows:
• risk < 20
• risk < 28
• risk < 33
• risk < 40
• risk ? 40